334 GOTE TURESSON 
each confined to a certain farm house or a certain steamer as men- 
tioned by HaGepoorn (1921), etc. 
That the differentiation of the plant species into the different here- 
ditary habitat types discussed in the first part of this paper cannot 
be adequately accounted for by recourse to such sporadic variability 
preserved by chance isolation will be made clear in the following. On 
the contrary it will be established that these habitat types in all pro- 
bability represent definite genotypical responses of the plant species to 
definite habitat factors. The data bearing upon this question have 
been collected under four headings. 
J.. THE SPECIALIZATION OF THE HABITAT TYPESiae 
DOMINANT HABITAT FACTORS. 
The effort so often made by ecologists to interpret the habitus and 
the morphological details of a plant as adaptations has no doubt led to 
undue generalization of the facts. It is evidently the genotypical con- 
stitution of the plant which is the point of primary importance. It is 
a generalization of the facts to maintain that the prostrate habit of 
growth enables the prostrate forms of Atriplex sarcophyllum to live on 
the exposed Swedish west-coast while erect forms of the same species, 
because of their erectness, are expelled from this coast strip. Compa- 
rative physiological experiments with individuals differing genotypically 
from each other only as to the factor or factors responsible for the dif- 
ference of growth would probably yield some information as to this 
point. It must be remembered, however, that it is the sum total of the 
genes, the »Gesamtgenotypus» (JOHANNSEN, 1914), which doutbless de- 
termine the presence or absence of a certain form in a certain habitat. 
Thanks to their genotypical constitution — and not necessarily on 
account of certain phenotypical, morphological characteristics — these 
prostrates support life where the erects do not. In some cases, however, 
a special habitus seems so related to definite habitat factors that we are 
compelled to assume that it is the particular reaction type of a particular 
genotype that enables the organism in question to live in a certain habi- 
tat. Such is the case with some of the above-discussed Hieracium habitat 
types. The dune type of the arenacious fields differs from the type 
of the shifting dunes mainly in the extreme prostrateness of the stems. 
It grows well in the grassy plains and stationary dunes but avoids the 
shifting sand dunes, where it would probably be sand-covered and die 
(since the power of shoot-regeneration is very limited in this type, 
though it is on the contrary most pronounced in the type of the 
